Changes between Version 1 and Version 2 of WhileLoopCarryOptimizationStrategy


Ignore:
Timestamp:
Dec 20, 2009, 8:53:13 AM (10 years ago)
Author:
cameron
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • WhileLoopCarryOptimizationStrategy

    v1 v2  
    1212
    1313
    14 
     14For example consider the comment-CDATA-PI loop in source:proto/parabix2/parabix2.py@322#L168.
     15There are 14 logical operations within the loop body.  So performing them 64 bits at a time
     16rather than 128 costs an additional 14 operations.   But with superscalar execution of 3 logical
     17operations per cycle the actual cost should only be about 5 cycles.   However, the entire
     18while loop can now be processed using 64-bit operations uniformly, avoiding any 128-bit
     19to 64-bit movements.   There are 7 variables involved in advance/scan operations.   This
     20would cost 4 data movement operations each per 128-bit block, or 28 operations in all.